The primary defining feature of field computation is that it operates on an entire field in parallel. For example, operations that process a retinal image in parallel, or which generate a spatial or motor map in parallel, are clear examples of field computation. On the other hand, a process that generates one or a few scalar signals sequentially in time is not considered field computation (except in a degenerate or trivial sense). The point is not to have a clear and absolutely precise demarcation between field computation and non-field computation -- it is fundamentally a matter of degree -- but to distinguish field computation as a style of computation from computation that is scalar or low-dimensional. The operational criterion is the ability to apply continuous mathematics to the spatial distribution of quantity.
In this section we consider field operations, which are
commonly implemented by nonrecurrent or feed-forward
connections between brain areas.
That is, a pattern of activity 
over an area A at
time t causes a pattern of activity 
over an area B at a slightly later time t'.
More generally, activity pattern 
over region B
depends on earlier activity patterns 
over regions 
:
[ (t) = F[_1(t-_1),...,_n(t-_n)],
]
where 
are fixed delays.
Field operations may be classified as linear (including
multilinear) or nonlinear.